JOURNAL OF PATHOLOGY, VOL.
164: 9-1 5 (1991)
MORPHOLOGICAL EVIDENCE THAT A-CAM IS A MAJOR INTERCELLULAR ADHESION MOLECULE IN HUMAN KIDNEY L. R. BIDDLESTONE AND S. FLEMING
Depurtment of Pathology, University of Edinburgh, Edinburgh EH9 SAG, U.K. Received 20 August 1990 Accepted 14 November 1990
SUMMARY We have used immunocytochemistryto identify the major primary adhesion molecule of the cadherin class in human kidney. In frozen sections of kidney, A-CAM was detected using the monoclonal antibody GC 4 on the surface of renal tubular epithelial cells. Renal tubular epithelium did not express L-CAM. No cadherin reactivity was found on the glomerular epithelial cells. Cultured renal tubular epithelium was studied by immunofluorescence and immunogold methods. A-CAM was found at the contact points ofadjacent epithelial cells, the phenotype ofwhich was confirmed by the demonstration of cytokeratins using the antibody CAM 5.2. The A-CAM molecule in human kidney had an M , of 130 kD in Western blotting experiments. These results lead us to conclude that A-CAM is the major cadherin of adult human renal epithelium. KEY
WORDS-A-CAM,kidney, cell adhesion, cadherins. INTRODUCTION
of integrins and cadherins following tubular epithelial induction. The major primary intercellular Cellular adhesion is believed to be important in the adhesion molecule during this morphogenetic proregulation of cell proliferation, cyto-differentiation, cess is a member of the cadherin family, which in the morphogenesis, and in the maintenance of multi- rodent kidney is L-CAM or u v ~ m o r u l i n .Sub~ cellular structures. Morphological studies of cell- sequent stabilization of adhesion is achieved by the cell adhesion have identified structurally distinct synthesis and assembly of the several components of types of cell junction and have defined their relation- the desmosomes which then act as an initiation site ship to the cytoskeleton. These include the two for the assembly of the intermediate filament cytomajor adhesive junctions desmosomes, which are ~ k e l e t o n .The ~ induced cells now have a simple associated with cytokeratin intermediate filaments, epithelial phenotype and further cytodifferentiation and adherens junctions, which interact with actin can occur. In the adult, differentiated kidney tubumicrofilaments. Biochemical and immunochemical lar epithelial integrity is maintained by cadherins methods have been used to define the molecules and stabilized by d e s m o s ~ m e s . ~ involved in cellkcell and cell-substratum adhesion. The cadherins, which act as primary adhesion These can be grouped into three families based on molecules in renal epithelial differentiation, are structural and functional criteria-the cadherins, membrane glycoproteins and mediate calciumintegrins, and immunoglobulin gene superfamily.2 dependent intercellular adhesion by homophilic During the differentiation of the renal epithelium binding. Study of tissue distribution and immunofrom the metanephric blastema, there are changes in logical specificities has led to the identification of the expression of these adhesion molecule^.^ In three subclasses ofcadherins called E-cadherins (epimurine kidney, there is loss of the immunoglobulin thelial cadherins), N-cadherins (neural cadherins), supergene family member N-CAM and expression and P-cadherins (placental cadherins).' In this study we provide morphological evidence Addressee for correspondence: Dr L. R. Biddlestone, Departthat the major primary cadherin involved in interment of Pathology, University of Edinburgh, Edinburgh cellular adhesion in human renal epithelium is not EH9 8AG, U.K.
'
0022-341 719 1/010009-07 $05.00 0 1991 by John Wiley & Sons, Ltd.
10
L. R. BIDDLESTONE AND S. FLEMING
L-CAM, as in rodent kidney, but A-CAM, which is a member of the N-cadherin subclass. We have used a mouse monoclonal antibody which recognizes the adhesive domain of A-CAM, the region most likely to show interspecies conservation, and a rat monoclonal antibody to L-CAM for immunolocalization of these molecules in human kidney and cultured human renal epithelial cells. MATERIALS AND METHODS Tissue sections
Blocks of morphological normal renal cortex were taken from fresh human nephrectomy specimens and snap-frozen in liquid nitrogen prior to storage at - 70°C. Serial 5 pm frozen sections were cut and mounted on glass slides for immunocytochemistry. Renal epithelial cell culture
Human renal epithelial cells were isolated by enzymatic dissociation using a modification of the method described by Kempson et al.’ Adult human renal cortex was obtained from fresh nephrectomy specimens and the cortical tissue cut into 1-2 mm cubes in phosphate-buffered saline (PBS). The tissue fragments were then incubated for 1 h at 37°C in a solution of 0.1 per cent type IV collagenase (Sigma) and 0.1 per cent trypsin (16 000 BAAE U/mg porcine pancreas, Sigma) in PBS. After incubation the resulting cell suspension was mixed with twice the volume of 0.16 per cent trypsin inhibitor (chicken eggwhite, Sigma) and centrifuged at 1900 rpm for 10 min. The supernatant was discarded and the cell pellet resuspended in Waymouth medium supplemented with 5 per cent fetal calf serum, penicillin (100 Ujml), and streptomycin (0. I mgiml). The cells were seeded to both plastic and glass coverslips in culture grade Petri dishes. Cultures were incubated at 37°C in a humidified atmosphere of 5 per cent CO, in air and the medium was changed at 2-day intervals. Cultures at near confluency were fixed either in acetone for 5-10 min at room temperature for immunofluorescence or in 3 per cent paraformaldehyde with 0.5 per cent Triton-X100 at 4°C for 15-20 min for immunogold staining. A n tibodies
A mouse monoclonal antibody, GC-4 (Sigma), raised against chick cardiac muscle intercalated discs, which recognizes the adhesive domain of A-
CAM was used diluted 1:50 for immunofluorescence and 1:20 for immunogold staining. A rat monoclonal antibody, DECMA- 1 (Sigma), to L-CAM (raised against a mouse embryonal carcinoma cell line) was used at a concentration of 1 :200 for immunofluorescence. CAM 5.2 (Becton Dickinson), a mousemonoclonal antibody labelling low relative molecular weight keratins, was used diluted 1:5 for both immunofluorescence and immunogold visualization techniques. A monoclonal antibody, DG 3.10 (Boehringer), to desmoglein, a 165 kD glycoprotein present in both the extracellular space and desmosoma1 plaque, was used at a concentration of 1 :4 for immunofluorescence. Immunojuorescence
Five pm frozen sections of normal human kidney and renal epithelial cell monolayers on glass coverslips were stained using a standard indirect immunofluorescence protocol with a 30 min incubation at room temperature for both primary and secondary antibodies. The secondary antibodies used were sheep-anti-mouse and rabbit-anti-rat fluorescein conjugates. Frozen sections and cultured cell monolayers were stained using this method for A-CAM, L-CAM, desmoglein-1, and cytokeratins (CAM 5.2). Cultured mouse glomerular epithelial cells provided a positive control for L-CAM and for desmoglein 1. Immunofluorescence specimens were viewed on a Zeiss laser scanning microscope with fluorescence excited by a 488 nm argon laser. Immunogold with silver enhancement visualization
Cultured cell monolayers were stained for ACAM and cytokeratins using a 30 min incubation with the primary antibodies and a 3 h incubation with the secondary antibody, a goat-anti-mouse 5 nm particle size colloidal gold conjugate (Auroprobe Janssen) at a dilution of 3:100, at room temperature. Silver amplification was performed using the Intense M kit (Janssen) followed by nuclear counterstaining with haematoxylin. Western blotting
A protein extract of human kidney was prepared by incubation of finely chopped renal cortical tissue with a urea-based extraction buffer ( 8 urea ~ in 50 mM Tris-HC1, 10 per cent 2-mercaptoethanol, 1 0 m ~PMSF, and 2 . 5 m ~EDTA) overnight at
A-CAM IN HUMAN KIDNEY
Fig. I-Reactivity for A-CAM is seen 011 the basolateral aspect of the renal tubular epithelial cells. N o reactivity is seen in the interstitium or on blood vessels. Immunofluorescence. GC4
room temperature. After extraction the solid material was separated by centrifugation and the supernatant stored at - 20°C. The extracted protein was diluted 1:4 in the electrophoresis sample buffer [8 M urea, 10 mM EDTA, 10 per cent (v/v) 2-mercaptoethanol, 5 per cent (w/v) SDS, and 0.04 per cent (w/v) bromophenol blue in 50mM Tris-HCI, pH 7.41 and separated by SDSPAGE using 12.5 per cent polyacrylamide gels and a solid Tris-Tricine discontinuous buffer system (Pharmacia Phast System). Standard molecular weight markers (Sigma) prepared in the same sample buffer were run on the same gel. Proteins were transferred to nitrocellulose by diffusion blotting at 70°C. Following transfer the nitrocellulose paper was washed in Tris-buffered saline (TBS), and blocking of free protein binding sites was carried out by soaking in TBS containing 5 per cent dried milk powder and 0.1 per cent Tween 20 overnight at 4°C. The nitrocellulose was washed in TBS containing 0.1 per cent (v/v) Tween 20 and 10 per cent (v/v) normal sheep serum (NSS) and then incubated with anti-A-CAM diluted 1 : l O in the TBS-Tween 20NSS solution for 1 h at room temperature. Visualization was performed using an avidin/biotin/alkaline phosphatase technique and the gel stained using
kenacid blue. Negative control experiments were performed in which normal mouse serum replaced the monoclonal antibody.
RESULTS Frozen sections of normal kidney The frozen sections stained using indirect immunofluorescence for cytokeratin showed cytoplasmic reactivity in the tubular epithelium and Bowman’s capsule but not in glomerular podocytes. Anti-desmoglein gave a punctate pattern of membrane staining on the lateral aspect at the apex of the tubular cells. The podocytes were negative for desmoglein. A-CAM reactivity was present on the basolateral aspect of the tubular epithelial cells throughout the length of the nephron (Fig. I ) but no reactivity was seen in the glomeruli, blood vessels, or mesangium. No L-CAM reactivity was present in the sections of normal human kidney. Cell cultures Cell attachment to the glass or plastic substratum was seen within 24 h and cultures reached subconfluence at 5-7 days. In the monolayers, large polygonal cells with an epithelial morphology were the
12
L. R. BIDDLESTONE AND S. FLEMING
Fig. 2-Cultured renal tubular epithelial cells show a fibrillary pattern of staining with CAM 5.2 confirming their epithelial phenotype. Imrnunofluorescence, CAM 5.2
predominant cell type but some spindle-shaped cells with long cytoplasmic processes were also present. Immunofluorescence staining for CAM 5.2 showed a fibrillary pattern of cytoplasmic positivity in the polygonal cells demonstrating the presence of a cytokeratin cytoskeleton and confirming the epithelial phenotype of these cultured cells (Fig. 2). Staining for desmoglein gave a punctate pattern of membrane positivity along the intercellular borders. The spindle-shaped cells showed no such immunoreactivity . A-CAM reactivity was seen along cell membranes at the intercellular border (Fig. 3), but was not present on the outer, free cell borders in subconfluent cultures. No L-CAM positivity was seen in the monolayers. L-CAM and desmoglein 1 reactivity was seen in the positive control experiments. Staining using the immunogold-silver technique for A-CAM and cytokeratin gave patterns of reactivity essentially similar to those seen with immunofluorescence. A-CAM reactivity was present along the cell membranes, and the points of cell-cell adhesion were seen as A-CAM-positive intercellular
contacts between adjacent cells (Fig. 4). As before, no A-CAM reactivity was present on outermost cell borders in subconfluent cultures. Westernblot ana[ysjs Immunoblot analysis of human kidney protein extract revealed an A-CAM-positive band with an apparent molecular weight of 130 kD (Fig. 5). A few additional weak bands were present at lower molecular weights which were also present in the negative controls.
DISCUSSION Cadherins are cell adhesion molecules involved in calcium-dependent cellkell adhesion. They were identified by screening antibodies against cell surface components for those that disrupt or inhibit calciumdependent cell aggregation. The subdivision of the cadherins was initially based on reactivity with a series of monoclonal antibodies raised against different tissues by Takeichi:6 ECCD- 1 (epithelial
A-CAM IN HUMAN KIDNEY
13
Fig. 3-A-CAM immunoreactivity is seen in the cell membranes of adjacent cells in the renal tubular epithelial cell monolayers. Immunofluorescence, GC4
cadherin), NCD-I and -2 (neural cadherin), and PCD- 1 (placental cadherin). Gene sequencing data have subsequently confirmed this subdivision. L-CAM and A-CAM are found in adherens junctions in different tissues and species. Transfection experiments show that the expression of E-cadherin, N-cadherin, and P-cadherin is associated with the formation of junctional structures and the development of an epithelioid phenotype.'-'' Studies of avian and murine embryogenesis suggest a major role for the cadherins in morphogenesis. The homophilic nature of cadherin interactions and the spatiotemporally regulated expression of different cadherins are important in cell sorting, separation of cell layers, and cell aggregation.6 A-CAM (adherens junction-specific cell adhesion molecule) belongs to the N-cadherin-like group of cadherins, and was first described in chicken cardiac muscle intercalated discs." Subsequent studies using chicken cardiac muscle and cultured lens epithelium showed that this 135 kD membrane glycoprotein is involved in calcium-dependent cellkell adhesion and is localized at adherens junctions." Incubation of cells with EDTA results in disruption of these cell junctions, whilst restoration of the calcium concentration in the culture medium is
followed by reformation of the cellkell adhesions.'* The presence of Fab fragments of anti-A-CAM during the calcium recovery phase inhibited the reformation of adherens junctions and was associated with disruption of the microfilament organization with fragmentation of actin bundles in the cytopIasm.I2 The tissue distribution of A-CAM has been studied mainly in avian tissues and some cultured cell lines. A-CAM has been described in chicken neural crest cells, brain, the basal layer of skin and cornea, cardiac muscle and lens epithelium, and in cultured MDBK cells. During avian renal development, ACAM is transiently expressed on the developing renal tubules, later disappearing along a proximodistal gradient coincident with the appearance of L-CAM, which is the major adhesion molecule in avian adult renal tubular epithelium." L-CAM, also known as uvomorulin, a 124 kD glycoprotein, belongs to the E-cadherin group of cadherins and was first identified in chicken hepatocytes.I4Embryological studies in chicken and mouse have shown that during renal development the expression of L-CAM is correlated with the differentiated epithelial phenotype. During neurulation, L-CAM disappears from the neural ectoderm but is
14
L. R. BIDDLESTONE AND S . FLEMING
Fig. +An immunogold specimen of cultured tubular epithelium shows A-CAM reactivity at the contact regions (arrows) of the cell membranes of adjacent cells. Immunogold-silver, GC4
still expressed in all endodermal cells, and in other ectodermal cells as long as they differentiate into epithelial It has been reported that A-CAM is absent from adult intestinal epithelium, hepatocytes, and kidney tubules. The tissue distributions of L-CAM and ACAM in adult and embryo avian and murine tissues are different and complementary, and it has been postulated that the cell adhesion molecule in adherens junctions is either L-CAM or A-CAM depending on the cell type.6 Thus, many of the descriptions of tissue distribution and functional studies of cadherin-mediated cell adhesion have been based on animal studies. There are no data regarding the distribution of ACAM in human tissue. In order to identify the major cadherin in human kidney, we have studied the distributions of A-CAM and L-CAM in adult human kidney and in cultured renal epithelial cells. In normal adult human kidney, we were unable to demonstrate any L-CAM reactivity, whereas striking A-CAM positivity was present along the basolateral aspect of the tubular epithelial cells. The significance of staining on the basal aspect of the
cells in tissue sections is not clear and the presence of A-CAM in the basal aspect of the tubular cells was not found in adult tubular cells in the cell culture experiments. The basal reactivity may represent lateral membrane staining in the complex interdigitation of tubular epithelial cells in intact tissue, as has previously been observed in investigations of desmosomal antigens in human k i d n e ~ . ~ The observed pattern of A-CAM reactivity in cultured human renal epithelial cells is entirely consistent with its role as a homotypic intercellular adhesion molecule. Membrane staining was seen along intercellular borders but not on the free outermost cell borders in subconfluent cultures. The localized sites of cellkell adhesion could be seen in the immunogold-silver preparations as A-CAMpositive intercellular contacts. Immunoblot analysis of human kidney protein extract showed an A-CAM-positive protein band with an apparent molecular weight of 130 kD, which is consistent with previous studies. The original description of A-CAM isolated from chicken cardiac muscle described a 135 kD glycoprotein" and subsequent Western blot analysis of chick
15
A-CAM IN H U M A N K I D N E Y
CONTROL
A-CAM
ACKNOWLEDGEMENTS
We wish to thank Shirley Dods for excellent technical assistance. This work was supported by the Scottish Hospitals Endowment Research Trust and the Scottish Home and Health Department.
-130 REFERENCES 1. Boyer B, Thiery JP. Epithelial cell adhesion mechanisms. J Mernhr Biol 1989: 112: 97-108.
IMMUNOBLOT
KIDNEY
Fig. 5-An A-CAM-positive band at M , 130 kD is seen in this Western blotting experiment performed with protein extracted from whole human kidney
embryo heart and lens showed an A-CAM-positive band with an apparent molecular weight of 130 kD.I5 In conclusion, we have shown that A-CAM, and not L-CAM, is the major cadherin in adult human kidney. We believe this to be the first description of A-CAM in human epithelium.
2. Fleming S. Cellular functions of adhesion molecules. J Puihnl 1990; 161: 189-190. 3. Ekblom P. Developmentally regulated conversion of mesenchyme to epithelium. Fuseh J 1989; 3 2141-2150. 4. Vestweber D, Kemler R. Ekblom P. Cell adhesion molecule uvomorulinduring kidney development. Drv B i d 1985: 112: 213-221. 5. Garrod DR, Fleming S. Early expression of desmosomal components during kidney tubule morphogenesis in human and murine embryos. Derelopmeni 1990; 108 313-322. 6. Takeichi M . The cadherins: cell-cell adhesion molecules controlling animal morphogenesis. Developmenr 1988; 102: 639-655. 7. Kempson SA, McAteer JA, Al-Mahrouq HA, Dousa TP, Dougherty GS, Evan AP. Proximal tubule characteristics of cultured human renal cortex epithelium. J Lab Clin Med 1989; 113 285-296. 8. Hatta K , Nose A, Nagafuchi A, Takeichi M . Cloning and expression of cDNA encoding a neural calcium-dependent cell adhesion molecule: its identity in the cadherin gene family. J Cell Bid 1988: 106: 873-88 I . 9. Mege R-M, Matsuzaki F, Gallin WJ, Goldberg JI, Cunningham BA, Edelman G M . Construction of epithelioid sheets by transfection of mouse sarcoma cells with cDNAs for chick cell adhesion molecules. Proc Nnrl AcudScr USA 1988; 8 5 7272-7278. 10. Matsuraki F, Mege R-M, Jaffe SH, et nl cDNAs of cell adhesion molecules of different specificity induced changes in cell shape and border formation in cultured s180 cells. J Cvll Biol 1990; 1 1 0 1239-1 252. 1 I . Volk T, Geiger B. A 135-kD membrane protein of intercellular adherensjunctions. EMBO J 1984; 3 2249-2260. 12. Volk T, Geiger B. A-CAM:a 135-kD receptor of intercellular adherens junctions. ii. Antibody mediated modulation of junction formation. J CellBiol1986; 103: 1451-1464. 13. Hatta K, Takagi S, Fujisawa H , Takeichi M . Spatial and temporal expression pattern of N-cadherin cell adhesion molecules correlated with morphogenetic processes ofchicken embryos. Dev B i d 1987; I 2 0 215-227. 14. Bertolotti R, Rutishauser U, Edelman G M . A cell surface molecule involved in aggregation of embryonic liver cells. Pro< N u t / Acad Sci USA 1980; 774831 4835 15. Duband J-L, Volberg T, Sabanay I, Thiery JP, Geiger 8. Spatial and temporal distribution of the adherens junction associated adhesion molecule A-CAM during embryogenesis. Developpmenr 1988; 103 325-344.
164: 9-1 5 (1991)
MORPHOLOGICAL EVIDENCE THAT A-CAM IS A MAJOR INTERCELLULAR ADHESION MOLECULE IN HUMAN KIDNEY L. R. BIDDLESTONE AND S. FLEMING
Depurtment of Pathology, University of Edinburgh, Edinburgh EH9 SAG, U.K. Received 20 August 1990 Accepted 14 November 1990
SUMMARY We have used immunocytochemistryto identify the major primary adhesion molecule of the cadherin class in human kidney. In frozen sections of kidney, A-CAM was detected using the monoclonal antibody GC 4 on the surface of renal tubular epithelial cells. Renal tubular epithelium did not express L-CAM. No cadherin reactivity was found on the glomerular epithelial cells. Cultured renal tubular epithelium was studied by immunofluorescence and immunogold methods. A-CAM was found at the contact points ofadjacent epithelial cells, the phenotype ofwhich was confirmed by the demonstration of cytokeratins using the antibody CAM 5.2. The A-CAM molecule in human kidney had an M , of 130 kD in Western blotting experiments. These results lead us to conclude that A-CAM is the major cadherin of adult human renal epithelium. KEY
WORDS-A-CAM,kidney, cell adhesion, cadherins. INTRODUCTION
of integrins and cadherins following tubular epithelial induction. The major primary intercellular Cellular adhesion is believed to be important in the adhesion molecule during this morphogenetic proregulation of cell proliferation, cyto-differentiation, cess is a member of the cadherin family, which in the morphogenesis, and in the maintenance of multi- rodent kidney is L-CAM or u v ~ m o r u l i n .Sub~ cellular structures. Morphological studies of cell- sequent stabilization of adhesion is achieved by the cell adhesion have identified structurally distinct synthesis and assembly of the several components of types of cell junction and have defined their relation- the desmosomes which then act as an initiation site ship to the cytoskeleton. These include the two for the assembly of the intermediate filament cytomajor adhesive junctions desmosomes, which are ~ k e l e t o n .The ~ induced cells now have a simple associated with cytokeratin intermediate filaments, epithelial phenotype and further cytodifferentiation and adherens junctions, which interact with actin can occur. In the adult, differentiated kidney tubumicrofilaments. Biochemical and immunochemical lar epithelial integrity is maintained by cadherins methods have been used to define the molecules and stabilized by d e s m o s ~ m e s . ~ involved in cellkcell and cell-substratum adhesion. The cadherins, which act as primary adhesion These can be grouped into three families based on molecules in renal epithelial differentiation, are structural and functional criteria-the cadherins, membrane glycoproteins and mediate calciumintegrins, and immunoglobulin gene superfamily.2 dependent intercellular adhesion by homophilic During the differentiation of the renal epithelium binding. Study of tissue distribution and immunofrom the metanephric blastema, there are changes in logical specificities has led to the identification of the expression of these adhesion molecule^.^ In three subclasses ofcadherins called E-cadherins (epimurine kidney, there is loss of the immunoglobulin thelial cadherins), N-cadherins (neural cadherins), supergene family member N-CAM and expression and P-cadherins (placental cadherins).' In this study we provide morphological evidence Addressee for correspondence: Dr L. R. Biddlestone, Departthat the major primary cadherin involved in interment of Pathology, University of Edinburgh, Edinburgh cellular adhesion in human renal epithelium is not EH9 8AG, U.K.
'
0022-341 719 1/010009-07 $05.00 0 1991 by John Wiley & Sons, Ltd.
10
L. R. BIDDLESTONE AND S. FLEMING
L-CAM, as in rodent kidney, but A-CAM, which is a member of the N-cadherin subclass. We have used a mouse monoclonal antibody which recognizes the adhesive domain of A-CAM, the region most likely to show interspecies conservation, and a rat monoclonal antibody to L-CAM for immunolocalization of these molecules in human kidney and cultured human renal epithelial cells. MATERIALS AND METHODS Tissue sections
Blocks of morphological normal renal cortex were taken from fresh human nephrectomy specimens and snap-frozen in liquid nitrogen prior to storage at - 70°C. Serial 5 pm frozen sections were cut and mounted on glass slides for immunocytochemistry. Renal epithelial cell culture
Human renal epithelial cells were isolated by enzymatic dissociation using a modification of the method described by Kempson et al.’ Adult human renal cortex was obtained from fresh nephrectomy specimens and the cortical tissue cut into 1-2 mm cubes in phosphate-buffered saline (PBS). The tissue fragments were then incubated for 1 h at 37°C in a solution of 0.1 per cent type IV collagenase (Sigma) and 0.1 per cent trypsin (16 000 BAAE U/mg porcine pancreas, Sigma) in PBS. After incubation the resulting cell suspension was mixed with twice the volume of 0.16 per cent trypsin inhibitor (chicken eggwhite, Sigma) and centrifuged at 1900 rpm for 10 min. The supernatant was discarded and the cell pellet resuspended in Waymouth medium supplemented with 5 per cent fetal calf serum, penicillin (100 Ujml), and streptomycin (0. I mgiml). The cells were seeded to both plastic and glass coverslips in culture grade Petri dishes. Cultures were incubated at 37°C in a humidified atmosphere of 5 per cent CO, in air and the medium was changed at 2-day intervals. Cultures at near confluency were fixed either in acetone for 5-10 min at room temperature for immunofluorescence or in 3 per cent paraformaldehyde with 0.5 per cent Triton-X100 at 4°C for 15-20 min for immunogold staining. A n tibodies
A mouse monoclonal antibody, GC-4 (Sigma), raised against chick cardiac muscle intercalated discs, which recognizes the adhesive domain of A-
CAM was used diluted 1:50 for immunofluorescence and 1:20 for immunogold staining. A rat monoclonal antibody, DECMA- 1 (Sigma), to L-CAM (raised against a mouse embryonal carcinoma cell line) was used at a concentration of 1 :200 for immunofluorescence. CAM 5.2 (Becton Dickinson), a mousemonoclonal antibody labelling low relative molecular weight keratins, was used diluted 1:5 for both immunofluorescence and immunogold visualization techniques. A monoclonal antibody, DG 3.10 (Boehringer), to desmoglein, a 165 kD glycoprotein present in both the extracellular space and desmosoma1 plaque, was used at a concentration of 1 :4 for immunofluorescence. Immunojuorescence
Five pm frozen sections of normal human kidney and renal epithelial cell monolayers on glass coverslips were stained using a standard indirect immunofluorescence protocol with a 30 min incubation at room temperature for both primary and secondary antibodies. The secondary antibodies used were sheep-anti-mouse and rabbit-anti-rat fluorescein conjugates. Frozen sections and cultured cell monolayers were stained using this method for A-CAM, L-CAM, desmoglein-1, and cytokeratins (CAM 5.2). Cultured mouse glomerular epithelial cells provided a positive control for L-CAM and for desmoglein 1. Immunofluorescence specimens were viewed on a Zeiss laser scanning microscope with fluorescence excited by a 488 nm argon laser. Immunogold with silver enhancement visualization
Cultured cell monolayers were stained for ACAM and cytokeratins using a 30 min incubation with the primary antibodies and a 3 h incubation with the secondary antibody, a goat-anti-mouse 5 nm particle size colloidal gold conjugate (Auroprobe Janssen) at a dilution of 3:100, at room temperature. Silver amplification was performed using the Intense M kit (Janssen) followed by nuclear counterstaining with haematoxylin. Western blotting
A protein extract of human kidney was prepared by incubation of finely chopped renal cortical tissue with a urea-based extraction buffer ( 8 urea ~ in 50 mM Tris-HC1, 10 per cent 2-mercaptoethanol, 1 0 m ~PMSF, and 2 . 5 m ~EDTA) overnight at
A-CAM IN HUMAN KIDNEY
Fig. I-Reactivity for A-CAM is seen 011 the basolateral aspect of the renal tubular epithelial cells. N o reactivity is seen in the interstitium or on blood vessels. Immunofluorescence. GC4
room temperature. After extraction the solid material was separated by centrifugation and the supernatant stored at - 20°C. The extracted protein was diluted 1:4 in the electrophoresis sample buffer [8 M urea, 10 mM EDTA, 10 per cent (v/v) 2-mercaptoethanol, 5 per cent (w/v) SDS, and 0.04 per cent (w/v) bromophenol blue in 50mM Tris-HCI, pH 7.41 and separated by SDSPAGE using 12.5 per cent polyacrylamide gels and a solid Tris-Tricine discontinuous buffer system (Pharmacia Phast System). Standard molecular weight markers (Sigma) prepared in the same sample buffer were run on the same gel. Proteins were transferred to nitrocellulose by diffusion blotting at 70°C. Following transfer the nitrocellulose paper was washed in Tris-buffered saline (TBS), and blocking of free protein binding sites was carried out by soaking in TBS containing 5 per cent dried milk powder and 0.1 per cent Tween 20 overnight at 4°C. The nitrocellulose was washed in TBS containing 0.1 per cent (v/v) Tween 20 and 10 per cent (v/v) normal sheep serum (NSS) and then incubated with anti-A-CAM diluted 1 : l O in the TBS-Tween 20NSS solution for 1 h at room temperature. Visualization was performed using an avidin/biotin/alkaline phosphatase technique and the gel stained using
kenacid blue. Negative control experiments were performed in which normal mouse serum replaced the monoclonal antibody.
RESULTS Frozen sections of normal kidney The frozen sections stained using indirect immunofluorescence for cytokeratin showed cytoplasmic reactivity in the tubular epithelium and Bowman’s capsule but not in glomerular podocytes. Anti-desmoglein gave a punctate pattern of membrane staining on the lateral aspect at the apex of the tubular cells. The podocytes were negative for desmoglein. A-CAM reactivity was present on the basolateral aspect of the tubular epithelial cells throughout the length of the nephron (Fig. I ) but no reactivity was seen in the glomeruli, blood vessels, or mesangium. No L-CAM reactivity was present in the sections of normal human kidney. Cell cultures Cell attachment to the glass or plastic substratum was seen within 24 h and cultures reached subconfluence at 5-7 days. In the monolayers, large polygonal cells with an epithelial morphology were the
12
L. R. BIDDLESTONE AND S. FLEMING
Fig. 2-Cultured renal tubular epithelial cells show a fibrillary pattern of staining with CAM 5.2 confirming their epithelial phenotype. Imrnunofluorescence, CAM 5.2
predominant cell type but some spindle-shaped cells with long cytoplasmic processes were also present. Immunofluorescence staining for CAM 5.2 showed a fibrillary pattern of cytoplasmic positivity in the polygonal cells demonstrating the presence of a cytokeratin cytoskeleton and confirming the epithelial phenotype of these cultured cells (Fig. 2). Staining for desmoglein gave a punctate pattern of membrane positivity along the intercellular borders. The spindle-shaped cells showed no such immunoreactivity . A-CAM reactivity was seen along cell membranes at the intercellular border (Fig. 3), but was not present on the outer, free cell borders in subconfluent cultures. No L-CAM positivity was seen in the monolayers. L-CAM and desmoglein 1 reactivity was seen in the positive control experiments. Staining using the immunogold-silver technique for A-CAM and cytokeratin gave patterns of reactivity essentially similar to those seen with immunofluorescence. A-CAM reactivity was present along the cell membranes, and the points of cell-cell adhesion were seen as A-CAM-positive intercellular
contacts between adjacent cells (Fig. 4). As before, no A-CAM reactivity was present on outermost cell borders in subconfluent cultures. Westernblot ana[ysjs Immunoblot analysis of human kidney protein extract revealed an A-CAM-positive band with an apparent molecular weight of 130 kD (Fig. 5). A few additional weak bands were present at lower molecular weights which were also present in the negative controls.
DISCUSSION Cadherins are cell adhesion molecules involved in calcium-dependent cellkell adhesion. They were identified by screening antibodies against cell surface components for those that disrupt or inhibit calciumdependent cell aggregation. The subdivision of the cadherins was initially based on reactivity with a series of monoclonal antibodies raised against different tissues by Takeichi:6 ECCD- 1 (epithelial
A-CAM IN HUMAN KIDNEY
13
Fig. 3-A-CAM immunoreactivity is seen in the cell membranes of adjacent cells in the renal tubular epithelial cell monolayers. Immunofluorescence, GC4
cadherin), NCD-I and -2 (neural cadherin), and PCD- 1 (placental cadherin). Gene sequencing data have subsequently confirmed this subdivision. L-CAM and A-CAM are found in adherens junctions in different tissues and species. Transfection experiments show that the expression of E-cadherin, N-cadherin, and P-cadherin is associated with the formation of junctional structures and the development of an epithelioid phenotype.'-'' Studies of avian and murine embryogenesis suggest a major role for the cadherins in morphogenesis. The homophilic nature of cadherin interactions and the spatiotemporally regulated expression of different cadherins are important in cell sorting, separation of cell layers, and cell aggregation.6 A-CAM (adherens junction-specific cell adhesion molecule) belongs to the N-cadherin-like group of cadherins, and was first described in chicken cardiac muscle intercalated discs." Subsequent studies using chicken cardiac muscle and cultured lens epithelium showed that this 135 kD membrane glycoprotein is involved in calcium-dependent cellkell adhesion and is localized at adherens junctions." Incubation of cells with EDTA results in disruption of these cell junctions, whilst restoration of the calcium concentration in the culture medium is
followed by reformation of the cellkell adhesions.'* The presence of Fab fragments of anti-A-CAM during the calcium recovery phase inhibited the reformation of adherens junctions and was associated with disruption of the microfilament organization with fragmentation of actin bundles in the cytopIasm.I2 The tissue distribution of A-CAM has been studied mainly in avian tissues and some cultured cell lines. A-CAM has been described in chicken neural crest cells, brain, the basal layer of skin and cornea, cardiac muscle and lens epithelium, and in cultured MDBK cells. During avian renal development, ACAM is transiently expressed on the developing renal tubules, later disappearing along a proximodistal gradient coincident with the appearance of L-CAM, which is the major adhesion molecule in avian adult renal tubular epithelium." L-CAM, also known as uvomorulin, a 124 kD glycoprotein, belongs to the E-cadherin group of cadherins and was first identified in chicken hepatocytes.I4Embryological studies in chicken and mouse have shown that during renal development the expression of L-CAM is correlated with the differentiated epithelial phenotype. During neurulation, L-CAM disappears from the neural ectoderm but is
14
L. R. BIDDLESTONE AND S . FLEMING
Fig. +An immunogold specimen of cultured tubular epithelium shows A-CAM reactivity at the contact regions (arrows) of the cell membranes of adjacent cells. Immunogold-silver, GC4
still expressed in all endodermal cells, and in other ectodermal cells as long as they differentiate into epithelial It has been reported that A-CAM is absent from adult intestinal epithelium, hepatocytes, and kidney tubules. The tissue distributions of L-CAM and ACAM in adult and embryo avian and murine tissues are different and complementary, and it has been postulated that the cell adhesion molecule in adherens junctions is either L-CAM or A-CAM depending on the cell type.6 Thus, many of the descriptions of tissue distribution and functional studies of cadherin-mediated cell adhesion have been based on animal studies. There are no data regarding the distribution of ACAM in human tissue. In order to identify the major cadherin in human kidney, we have studied the distributions of A-CAM and L-CAM in adult human kidney and in cultured renal epithelial cells. In normal adult human kidney, we were unable to demonstrate any L-CAM reactivity, whereas striking A-CAM positivity was present along the basolateral aspect of the tubular epithelial cells. The significance of staining on the basal aspect of the
cells in tissue sections is not clear and the presence of A-CAM in the basal aspect of the tubular cells was not found in adult tubular cells in the cell culture experiments. The basal reactivity may represent lateral membrane staining in the complex interdigitation of tubular epithelial cells in intact tissue, as has previously been observed in investigations of desmosomal antigens in human k i d n e ~ . ~ The observed pattern of A-CAM reactivity in cultured human renal epithelial cells is entirely consistent with its role as a homotypic intercellular adhesion molecule. Membrane staining was seen along intercellular borders but not on the free outermost cell borders in subconfluent cultures. The localized sites of cellkell adhesion could be seen in the immunogold-silver preparations as A-CAMpositive intercellular contacts. Immunoblot analysis of human kidney protein extract showed an A-CAM-positive protein band with an apparent molecular weight of 130 kD, which is consistent with previous studies. The original description of A-CAM isolated from chicken cardiac muscle described a 135 kD glycoprotein" and subsequent Western blot analysis of chick
15
A-CAM IN H U M A N K I D N E Y
CONTROL
A-CAM
ACKNOWLEDGEMENTS
We wish to thank Shirley Dods for excellent technical assistance. This work was supported by the Scottish Hospitals Endowment Research Trust and the Scottish Home and Health Department.
-130 REFERENCES 1. Boyer B, Thiery JP. Epithelial cell adhesion mechanisms. J Mernhr Biol 1989: 112: 97-108.
IMMUNOBLOT
KIDNEY
Fig. 5-An A-CAM-positive band at M , 130 kD is seen in this Western blotting experiment performed with protein extracted from whole human kidney
embryo heart and lens showed an A-CAM-positive band with an apparent molecular weight of 130 kD.I5 In conclusion, we have shown that A-CAM, and not L-CAM, is the major cadherin in adult human kidney. We believe this to be the first description of A-CAM in human epithelium.
2. Fleming S. Cellular functions of adhesion molecules. J Puihnl 1990; 161: 189-190. 3. Ekblom P. Developmentally regulated conversion of mesenchyme to epithelium. Fuseh J 1989; 3 2141-2150. 4. Vestweber D, Kemler R. Ekblom P. Cell adhesion molecule uvomorulinduring kidney development. Drv B i d 1985: 112: 213-221. 5. Garrod DR, Fleming S. Early expression of desmosomal components during kidney tubule morphogenesis in human and murine embryos. Derelopmeni 1990; 108 313-322. 6. Takeichi M . The cadherins: cell-cell adhesion molecules controlling animal morphogenesis. Developmenr 1988; 102: 639-655. 7. Kempson SA, McAteer JA, Al-Mahrouq HA, Dousa TP, Dougherty GS, Evan AP. Proximal tubule characteristics of cultured human renal cortex epithelium. J Lab Clin Med 1989; 113 285-296. 8. Hatta K , Nose A, Nagafuchi A, Takeichi M . Cloning and expression of cDNA encoding a neural calcium-dependent cell adhesion molecule: its identity in the cadherin gene family. J Cell Bid 1988: 106: 873-88 I . 9. Mege R-M, Matsuzaki F, Gallin WJ, Goldberg JI, Cunningham BA, Edelman G M . Construction of epithelioid sheets by transfection of mouse sarcoma cells with cDNAs for chick cell adhesion molecules. Proc Nnrl AcudScr USA 1988; 8 5 7272-7278. 10. Matsuraki F, Mege R-M, Jaffe SH, et nl cDNAs of cell adhesion molecules of different specificity induced changes in cell shape and border formation in cultured s180 cells. J Cvll Biol 1990; 1 1 0 1239-1 252. 1 I . Volk T, Geiger B. A 135-kD membrane protein of intercellular adherensjunctions. EMBO J 1984; 3 2249-2260. 12. Volk T, Geiger B. A-CAM:a 135-kD receptor of intercellular adherens junctions. ii. Antibody mediated modulation of junction formation. J CellBiol1986; 103: 1451-1464. 13. Hatta K, Takagi S, Fujisawa H , Takeichi M . Spatial and temporal expression pattern of N-cadherin cell adhesion molecules correlated with morphogenetic processes ofchicken embryos. Dev B i d 1987; I 2 0 215-227. 14. Bertolotti R, Rutishauser U, Edelman G M . A cell surface molecule involved in aggregation of embryonic liver cells. Pro< N u t / Acad Sci USA 1980; 774831 4835 15. Duband J-L, Volberg T, Sabanay I, Thiery JP, Geiger 8. Spatial and temporal distribution of the adherens junction associated adhesion molecule A-CAM during embryogenesis. Developpmenr 1988; 103 325-344.
